Laser galvanometer processing algorithm based on Bezier curve optimization of point set
-
摘要: 为了解决激光振镜加工中由于数据点集过于密集导致激光在单点停滞时间过长引起激光烧蚀问题,将原始数据点集经过稀疏化、曲线点集分段、三角二分法贝塞尔曲线拟合数据处理,转换成由少量直线和曲线构成的实体,在激光路径输出时,根据输出放大倍数的大小及曲线曲率大小对曲线进行整体的适应性插补,从而解决因放大倍数不同引起的数据点过密或过疏的问题,并通过对前后数据以不同的标刻分辨率进行标刻。结果表明,在保证拟合后的加工误差不超过2μm下,解决了数据过密引起的激光烧蚀问题;在高放大倍数下也具有平滑的加工效果,具有更高的加工效率和质量。此研究为高密集数据激光加工提供了参考,在高质量激光加工领域如激光精密刻蚀有良好的应用前景。Abstract: In order to solve the problem of laser ablation due to the long time laser stop at the single point caused by the excessive density of the data point set in laser galvanometer processing, the original data point set was processed by thinning, curve point set segmentation, and triangular dichotomy Bezier curve fitting data processing, which was converted into an entity composed of a few line and curve. When laser marking, the curve was adaptively interpolated according to the output resolution and curve curvature to solve the problem of too dense or too sparse data points caused by different magnification. By marking the original and processed data with different marking magnification, the results show that the fitted data eliminates the laser ablation phenomenon caused by over-density of the data under the guaranteed 2μm accuracy of the original data. It also has a smoother marking effect at high magnification, as well as a higher marking efficiency and quality. It provides a reference for high density laser processing and has a good application prospect in the field of high quality laser processing such as laser precision etching.
-
-
Table 1 Comparison of marking quality between original point set and fitted control points at different magnification
original point set fitted point set low magnification (36×) high magnification (72×) low magnification (36×) high magnification (72×) number of control point none none 40 40 mark point 1181 1181 207 288 fitting error/μm 0 0 1 2 processing time/ms 19 18 4 5 has laser ablation yes yes no no processing quality low low high high -
[1] LIU J J, YAN P, LI G Zh, et al. Laser processing technology[J]. Physics Bulletin, 2009(8): 54-56(in Chinese).
[2] XU K K. Research on high-speed galvanometer scanning laser 3-D printing control system[D]. Wuhan: Huazhong University of Science and Technology, 2017: 4-55(in Chinese).
[3] WANG B, LIU X D, LI J H, et al. Research on dot laser drilling location algorithm based on neighborhood features[J]. Laser Technology, 2019, 43(5): 1-6(in Chinese).
[4] LUO X, LI J, LUCAS M. Galvanometer scanning technology for laser additive manufacturing[J]. Proceedings of the SPIE, 2017, 10095: 1009512. DOI: 10.1117/12.2252973.full
[5] GAO X L. Binarization of image processing in laser engraving[J]. Laser Journal, 2004, 25(4): 76-78(in Chinese).
[6] WANG J P, LI Zh J, FAN X H. Analysis of laser marking system and process parameters[J]. Optics and Optoelectronic Technology, 2005, 3(3): 32-35(in Chinese). http://en.cnki.com.cn/Article_en/CJFDTOTAL-GXGD200503009.htm
[7] XU L P, HU B. A conversion method from JPG image to PLT file[J]. Laser Journal, 2005, 26(6): 50-51(in Chinese).
[8] LIU T L, HU B, YING H Sh, et al. Experimental research on laser marking error[J]. Laser Journal, 2007, 28(1): 33-34(in Chinese).
[9] LIAO P, YANG D Y, LIU R X. An optimization algorithm of truetype font contour line approximation in laser marking[J]. Laser Technology, 2016, 40(4): 483-486(in Chinese).
[10] HAN X L, LIU Sh J. Extension of cubic uniform B-spline curve[J]. Journal of Cofmputer Aided Design and Graphics, 2003, 15(5): 576-578(in Chinese).
[11] PIEGL L, TILLER W. The NURBS book[M]. 2th ed. New York, USA: Springer, 1997: 410-419.
[12] LIU M Z, GUO Q J, WANG S Q. Adaptive B-spline curve fitting based on regular asymptotic iterative approximation[J]. Journal of Graphics, 2018, 39(2): 287-294(in Chinese). http://www.zhangqiaokeyan.com/academic-journal-cn_journal-graphics_thesis/0201211797296.html
[13] SELINGER P. Potrace: A polygon-based tracing algorithm[DB/OL]. (2003-9-20)[2020-10-16]. http://potrace.sourceforge.net/potrace.pdf.
[14] YAO P P, FAN Zh, ZHANG S L. Improved Potrace jacquard image vectorization algorithm[J]. Sensors and Microsystems, 2014, 33(4): 125-127(in Chinese). http://en.cnki.com.cn/Article_en/CJFDTOTAL-CGQJ201404037.htm
[15] YU X J, PENG L Zh. A new algorithm for curve contour extraction in binary image[J]. Journal of Image and Graphics, 2002, A7(3): 272-275(in Chinese).
[16] ZHANG Ch C, SUN X M, ZHU T Y. Research on AutoCAD's DXF file format and its conversion interface[J]. Microcomputer Applications, 2001, 17(8): 54-55 (in Chinese). http://en.cnki.com.cn/Article_en/CJFDTOTAL-WXDY200108018.htm
[17] HETAL N F, AKASH B P, DIVYANG D P, et al. A review on approaches for handling bezier curves in CAD for manufacturing[J]. Procedia Engineering, 2014, 97: 1155-1166. DOI: 10.1016/j.proeng.2014.12.394
[18] YANG J. Research and implementation of key algorithms for forward-looking direct interpolation of NURBS curves in high-speed machining numerical control systems[D]. Hangzhou: Zhejiang University, 2019 : 4-32(in Chinese).
[19] XU Y M, WEN Sh Ch. Study on Bezier curve recursive algorithm[J]. Journal of Hengyang Normal University, 2007, 28 (6): 113-115(in Chinese). http://search.cnki.net/down/default.aspx?filename=HYSF200706028&dbcode=CJFD&year=2007&dflag=pdfdown
[20] KUANG Y, WU Y F, WU B, et al. Extraction and optimization of marking path in laser marking system[J]. Laser Technology, 2012, 36(1): 131-133(in Chinese). http://en.cnki.com.cn/Article_en/CJFDTOTAL-JGJS201201035.htm